File size: 25,158 Bytes
c341951 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 |
2023-10-06 09:54:44,470 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,471 Model: "SequenceTagger(
(embeddings): ByT5Embeddings(
(model): T5EncoderModel(
(shared): Embedding(384, 1472)
(encoder): T5Stack(
(embed_tokens): Embedding(384, 1472)
(block): ModuleList(
(0): T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
(relative_attention_bias): Embedding(32, 6)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(1-11): 11 x T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-06 09:54:44,471 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,472 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-06 09:54:44,472 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,472 Train: 1214 sentences
2023-10-06 09:54:44,472 (train_with_dev=False, train_with_test=False)
2023-10-06 09:54:44,472 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,472 Training Params:
2023-10-06 09:54:44,472 - learning_rate: "0.00016"
2023-10-06 09:54:44,472 - mini_batch_size: "4"
2023-10-06 09:54:44,472 - max_epochs: "10"
2023-10-06 09:54:44,472 - shuffle: "True"
2023-10-06 09:54:44,472 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,472 Plugins:
2023-10-06 09:54:44,472 - TensorboardLogger
2023-10-06 09:54:44,472 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 09:54:44,472 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,472 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 09:54:44,472 - metric: "('micro avg', 'f1-score')"
2023-10-06 09:54:44,473 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,473 Computation:
2023-10-06 09:54:44,473 - compute on device: cuda:0
2023-10-06 09:54:44,473 - embedding storage: none
2023-10-06 09:54:44,473 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,473 Model training base path: "hmbench-ajmc/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1"
2023-10-06 09:54:44,473 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,473 ----------------------------------------------------------------------------------------------------
2023-10-06 09:54:44,473 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 09:54:56,376 epoch 1 - iter 30/304 - loss 3.22742643 - time (sec): 11.90 - samples/sec: 288.60 - lr: 0.000015 - momentum: 0.000000
2023-10-06 09:55:07,887 epoch 1 - iter 60/304 - loss 3.21915042 - time (sec): 23.41 - samples/sec: 287.96 - lr: 0.000031 - momentum: 0.000000
2023-10-06 09:55:18,992 epoch 1 - iter 90/304 - loss 3.19773545 - time (sec): 34.52 - samples/sec: 282.12 - lr: 0.000047 - momentum: 0.000000
2023-10-06 09:55:30,310 epoch 1 - iter 120/304 - loss 3.13133843 - time (sec): 45.84 - samples/sec: 278.62 - lr: 0.000063 - momentum: 0.000000
2023-10-06 09:55:41,121 epoch 1 - iter 150/304 - loss 3.03631970 - time (sec): 56.65 - samples/sec: 274.31 - lr: 0.000078 - momentum: 0.000000
2023-10-06 09:55:51,674 epoch 1 - iter 180/304 - loss 2.92654304 - time (sec): 67.20 - samples/sec: 271.15 - lr: 0.000094 - momentum: 0.000000
2023-10-06 09:56:02,500 epoch 1 - iter 210/304 - loss 2.80021159 - time (sec): 78.03 - samples/sec: 270.63 - lr: 0.000110 - momentum: 0.000000
2023-10-06 09:56:13,664 epoch 1 - iter 240/304 - loss 2.66006911 - time (sec): 89.19 - samples/sec: 270.26 - lr: 0.000126 - momentum: 0.000000
2023-10-06 09:56:25,332 epoch 1 - iter 270/304 - loss 2.50546540 - time (sec): 100.86 - samples/sec: 269.65 - lr: 0.000142 - momentum: 0.000000
2023-10-06 09:56:37,599 epoch 1 - iter 300/304 - loss 2.33574461 - time (sec): 113.13 - samples/sec: 271.35 - lr: 0.000157 - momentum: 0.000000
2023-10-06 09:56:38,870 ----------------------------------------------------------------------------------------------------
2023-10-06 09:56:38,870 EPOCH 1 done: loss 2.3228 - lr: 0.000157
2023-10-06 09:56:46,110 DEV : loss 0.8970457911491394 - f1-score (micro avg) 0.0
2023-10-06 09:56:46,116 ----------------------------------------------------------------------------------------------------
2023-10-06 09:56:57,526 epoch 2 - iter 30/304 - loss 0.81144182 - time (sec): 11.41 - samples/sec: 269.45 - lr: 0.000158 - momentum: 0.000000
2023-10-06 09:57:09,002 epoch 2 - iter 60/304 - loss 0.75433976 - time (sec): 22.89 - samples/sec: 263.93 - lr: 0.000157 - momentum: 0.000000
2023-10-06 09:57:20,565 epoch 2 - iter 90/304 - loss 0.71555452 - time (sec): 34.45 - samples/sec: 263.96 - lr: 0.000155 - momentum: 0.000000
2023-10-06 09:57:32,831 epoch 2 - iter 120/304 - loss 0.66835125 - time (sec): 46.71 - samples/sec: 268.44 - lr: 0.000153 - momentum: 0.000000
2023-10-06 09:57:43,867 epoch 2 - iter 150/304 - loss 0.63094298 - time (sec): 57.75 - samples/sec: 265.47 - lr: 0.000151 - momentum: 0.000000
2023-10-06 09:57:55,696 epoch 2 - iter 180/304 - loss 0.57924068 - time (sec): 69.58 - samples/sec: 264.84 - lr: 0.000150 - momentum: 0.000000
2023-10-06 09:58:07,350 epoch 2 - iter 210/304 - loss 0.53846233 - time (sec): 81.23 - samples/sec: 262.91 - lr: 0.000148 - momentum: 0.000000
2023-10-06 09:58:19,675 epoch 2 - iter 240/304 - loss 0.51040511 - time (sec): 93.56 - samples/sec: 262.44 - lr: 0.000146 - momentum: 0.000000
2023-10-06 09:58:31,862 epoch 2 - iter 270/304 - loss 0.49514733 - time (sec): 105.74 - samples/sec: 261.92 - lr: 0.000144 - momentum: 0.000000
2023-10-06 09:58:43,426 epoch 2 - iter 300/304 - loss 0.47727700 - time (sec): 117.31 - samples/sec: 261.15 - lr: 0.000143 - momentum: 0.000000
2023-10-06 09:58:44,875 ----------------------------------------------------------------------------------------------------
2023-10-06 09:58:44,876 EPOCH 2 done: loss 0.4740 - lr: 0.000143
2023-10-06 09:58:53,032 DEV : loss 0.31851717829704285 - f1-score (micro avg) 0.4416
2023-10-06 09:58:53,040 saving best model
2023-10-06 09:58:53,870 ----------------------------------------------------------------------------------------------------
2023-10-06 09:59:06,320 epoch 3 - iter 30/304 - loss 0.26608500 - time (sec): 12.45 - samples/sec: 257.05 - lr: 0.000141 - momentum: 0.000000
2023-10-06 09:59:18,741 epoch 3 - iter 60/304 - loss 0.23335679 - time (sec): 24.87 - samples/sec: 257.38 - lr: 0.000139 - momentum: 0.000000
2023-10-06 09:59:30,077 epoch 3 - iter 90/304 - loss 0.22225252 - time (sec): 36.21 - samples/sec: 253.38 - lr: 0.000137 - momentum: 0.000000
2023-10-06 09:59:42,424 epoch 3 - iter 120/304 - loss 0.22829801 - time (sec): 48.55 - samples/sec: 256.38 - lr: 0.000135 - momentum: 0.000000
2023-10-06 09:59:53,935 epoch 3 - iter 150/304 - loss 0.22362351 - time (sec): 60.06 - samples/sec: 254.76 - lr: 0.000134 - momentum: 0.000000
2023-10-06 10:00:05,592 epoch 3 - iter 180/304 - loss 0.22215273 - time (sec): 71.72 - samples/sec: 255.33 - lr: 0.000132 - momentum: 0.000000
2023-10-06 10:00:17,824 epoch 3 - iter 210/304 - loss 0.21460363 - time (sec): 83.95 - samples/sec: 255.66 - lr: 0.000130 - momentum: 0.000000
2023-10-06 10:00:30,321 epoch 3 - iter 240/304 - loss 0.20673630 - time (sec): 96.45 - samples/sec: 256.78 - lr: 0.000128 - momentum: 0.000000
2023-10-06 10:00:41,730 epoch 3 - iter 270/304 - loss 0.20231564 - time (sec): 107.86 - samples/sec: 256.59 - lr: 0.000127 - momentum: 0.000000
2023-10-06 10:00:52,833 epoch 3 - iter 300/304 - loss 0.19777168 - time (sec): 118.96 - samples/sec: 256.91 - lr: 0.000125 - momentum: 0.000000
2023-10-06 10:00:54,278 ----------------------------------------------------------------------------------------------------
2023-10-06 10:00:54,278 EPOCH 3 done: loss 0.1959 - lr: 0.000125
2023-10-06 10:01:01,511 DEV : loss 0.18969886004924774 - f1-score (micro avg) 0.7057
2023-10-06 10:01:01,520 saving best model
2023-10-06 10:01:05,856 ----------------------------------------------------------------------------------------------------
2023-10-06 10:01:17,088 epoch 4 - iter 30/304 - loss 0.12053707 - time (sec): 11.23 - samples/sec: 270.96 - lr: 0.000123 - momentum: 0.000000
2023-10-06 10:01:28,458 epoch 4 - iter 60/304 - loss 0.13105014 - time (sec): 22.60 - samples/sec: 268.75 - lr: 0.000121 - momentum: 0.000000
2023-10-06 10:01:39,577 epoch 4 - iter 90/304 - loss 0.12897336 - time (sec): 33.72 - samples/sec: 266.85 - lr: 0.000119 - momentum: 0.000000
2023-10-06 10:01:50,991 epoch 4 - iter 120/304 - loss 0.12192808 - time (sec): 45.13 - samples/sec: 266.81 - lr: 0.000118 - momentum: 0.000000
2023-10-06 10:02:03,220 epoch 4 - iter 150/304 - loss 0.12197006 - time (sec): 57.36 - samples/sec: 271.13 - lr: 0.000116 - momentum: 0.000000
2023-10-06 10:02:14,426 epoch 4 - iter 180/304 - loss 0.11629911 - time (sec): 68.57 - samples/sec: 270.53 - lr: 0.000114 - momentum: 0.000000
2023-10-06 10:02:25,342 epoch 4 - iter 210/304 - loss 0.11116342 - time (sec): 79.48 - samples/sec: 270.01 - lr: 0.000112 - momentum: 0.000000
2023-10-06 10:02:36,770 epoch 4 - iter 240/304 - loss 0.10660005 - time (sec): 90.91 - samples/sec: 270.52 - lr: 0.000111 - momentum: 0.000000
2023-10-06 10:02:48,070 epoch 4 - iter 270/304 - loss 0.10527643 - time (sec): 102.21 - samples/sec: 269.97 - lr: 0.000109 - momentum: 0.000000
2023-10-06 10:02:59,578 epoch 4 - iter 300/304 - loss 0.10675985 - time (sec): 113.72 - samples/sec: 269.88 - lr: 0.000107 - momentum: 0.000000
2023-10-06 10:03:00,863 ----------------------------------------------------------------------------------------------------
2023-10-06 10:03:00,864 EPOCH 4 done: loss 0.1062 - lr: 0.000107
2023-10-06 10:03:08,009 DEV : loss 0.14084239304065704 - f1-score (micro avg) 0.8033
2023-10-06 10:03:08,018 saving best model
2023-10-06 10:03:12,371 ----------------------------------------------------------------------------------------------------
2023-10-06 10:03:23,771 epoch 5 - iter 30/304 - loss 0.06645253 - time (sec): 11.40 - samples/sec: 281.09 - lr: 0.000105 - momentum: 0.000000
2023-10-06 10:03:35,248 epoch 5 - iter 60/304 - loss 0.06334492 - time (sec): 22.88 - samples/sec: 274.19 - lr: 0.000103 - momentum: 0.000000
2023-10-06 10:03:46,693 epoch 5 - iter 90/304 - loss 0.07146608 - time (sec): 34.32 - samples/sec: 270.86 - lr: 0.000102 - momentum: 0.000000
2023-10-06 10:03:58,057 epoch 5 - iter 120/304 - loss 0.06442638 - time (sec): 45.68 - samples/sec: 268.61 - lr: 0.000100 - momentum: 0.000000
2023-10-06 10:04:09,037 epoch 5 - iter 150/304 - loss 0.06651026 - time (sec): 56.66 - samples/sec: 264.08 - lr: 0.000098 - momentum: 0.000000
2023-10-06 10:04:21,637 epoch 5 - iter 180/304 - loss 0.07204840 - time (sec): 69.26 - samples/sec: 266.08 - lr: 0.000096 - momentum: 0.000000
2023-10-06 10:04:32,999 epoch 5 - iter 210/304 - loss 0.07177304 - time (sec): 80.63 - samples/sec: 265.09 - lr: 0.000094 - momentum: 0.000000
2023-10-06 10:04:45,215 epoch 5 - iter 240/304 - loss 0.07011091 - time (sec): 92.84 - samples/sec: 265.16 - lr: 0.000093 - momentum: 0.000000
2023-10-06 10:04:56,839 epoch 5 - iter 270/304 - loss 0.07061563 - time (sec): 104.47 - samples/sec: 265.47 - lr: 0.000091 - momentum: 0.000000
2023-10-06 10:05:08,637 epoch 5 - iter 300/304 - loss 0.06762143 - time (sec): 116.26 - samples/sec: 264.50 - lr: 0.000089 - momentum: 0.000000
2023-10-06 10:05:09,823 ----------------------------------------------------------------------------------------------------
2023-10-06 10:05:09,823 EPOCH 5 done: loss 0.0676 - lr: 0.000089
2023-10-06 10:05:17,506 DEV : loss 0.14746029675006866 - f1-score (micro avg) 0.8037
2023-10-06 10:05:17,513 saving best model
2023-10-06 10:05:22,299 ----------------------------------------------------------------------------------------------------
2023-10-06 10:05:34,206 epoch 6 - iter 30/304 - loss 0.04698576 - time (sec): 11.90 - samples/sec: 261.58 - lr: 0.000087 - momentum: 0.000000
2023-10-06 10:05:45,667 epoch 6 - iter 60/304 - loss 0.05984385 - time (sec): 23.37 - samples/sec: 253.91 - lr: 0.000085 - momentum: 0.000000
2023-10-06 10:05:57,678 epoch 6 - iter 90/304 - loss 0.05653784 - time (sec): 35.38 - samples/sec: 257.09 - lr: 0.000084 - momentum: 0.000000
2023-10-06 10:06:09,777 epoch 6 - iter 120/304 - loss 0.04619937 - time (sec): 47.48 - samples/sec: 256.55 - lr: 0.000082 - momentum: 0.000000
2023-10-06 10:06:22,306 epoch 6 - iter 150/304 - loss 0.05426085 - time (sec): 60.01 - samples/sec: 257.01 - lr: 0.000080 - momentum: 0.000000
2023-10-06 10:06:34,087 epoch 6 - iter 180/304 - loss 0.05153527 - time (sec): 71.79 - samples/sec: 257.04 - lr: 0.000078 - momentum: 0.000000
2023-10-06 10:06:46,005 epoch 6 - iter 210/304 - loss 0.04893707 - time (sec): 83.70 - samples/sec: 255.79 - lr: 0.000077 - momentum: 0.000000
2023-10-06 10:06:57,998 epoch 6 - iter 240/304 - loss 0.05206313 - time (sec): 95.70 - samples/sec: 255.94 - lr: 0.000075 - momentum: 0.000000
2023-10-06 10:07:10,628 epoch 6 - iter 270/304 - loss 0.05177056 - time (sec): 108.33 - samples/sec: 255.76 - lr: 0.000073 - momentum: 0.000000
2023-10-06 10:07:22,152 epoch 6 - iter 300/304 - loss 0.05044602 - time (sec): 119.85 - samples/sec: 255.02 - lr: 0.000071 - momentum: 0.000000
2023-10-06 10:07:23,729 ----------------------------------------------------------------------------------------------------
2023-10-06 10:07:23,730 EPOCH 6 done: loss 0.0507 - lr: 0.000071
2023-10-06 10:07:31,724 DEV : loss 0.15308107435703278 - f1-score (micro avg) 0.8071
2023-10-06 10:07:31,733 saving best model
2023-10-06 10:07:36,056 ----------------------------------------------------------------------------------------------------
2023-10-06 10:07:47,550 epoch 7 - iter 30/304 - loss 0.04688673 - time (sec): 11.49 - samples/sec: 245.11 - lr: 0.000069 - momentum: 0.000000
2023-10-06 10:07:59,541 epoch 7 - iter 60/304 - loss 0.05174308 - time (sec): 23.48 - samples/sec: 252.60 - lr: 0.000068 - momentum: 0.000000
2023-10-06 10:08:11,552 epoch 7 - iter 90/304 - loss 0.05838398 - time (sec): 35.49 - samples/sec: 255.59 - lr: 0.000066 - momentum: 0.000000
2023-10-06 10:08:23,942 epoch 7 - iter 120/304 - loss 0.04573919 - time (sec): 47.89 - samples/sec: 258.72 - lr: 0.000064 - momentum: 0.000000
2023-10-06 10:08:36,137 epoch 7 - iter 150/304 - loss 0.04438194 - time (sec): 60.08 - samples/sec: 258.27 - lr: 0.000062 - momentum: 0.000000
2023-10-06 10:08:47,827 epoch 7 - iter 180/304 - loss 0.04129594 - time (sec): 71.77 - samples/sec: 257.35 - lr: 0.000061 - momentum: 0.000000
2023-10-06 10:08:59,370 epoch 7 - iter 210/304 - loss 0.04159775 - time (sec): 83.31 - samples/sec: 255.87 - lr: 0.000059 - momentum: 0.000000
2023-10-06 10:09:11,738 epoch 7 - iter 240/304 - loss 0.04032401 - time (sec): 95.68 - samples/sec: 255.99 - lr: 0.000057 - momentum: 0.000000
2023-10-06 10:09:23,519 epoch 7 - iter 270/304 - loss 0.04184823 - time (sec): 107.46 - samples/sec: 255.77 - lr: 0.000055 - momentum: 0.000000
2023-10-06 10:09:35,883 epoch 7 - iter 300/304 - loss 0.03910036 - time (sec): 119.83 - samples/sec: 256.10 - lr: 0.000054 - momentum: 0.000000
2023-10-06 10:09:37,347 ----------------------------------------------------------------------------------------------------
2023-10-06 10:09:37,347 EPOCH 7 done: loss 0.0400 - lr: 0.000054
2023-10-06 10:09:45,183 DEV : loss 0.15628251433372498 - f1-score (micro avg) 0.8181
2023-10-06 10:09:45,190 saving best model
2023-10-06 10:09:49,516 ----------------------------------------------------------------------------------------------------
2023-10-06 10:10:02,225 epoch 8 - iter 30/304 - loss 0.01383221 - time (sec): 12.71 - samples/sec: 262.38 - lr: 0.000052 - momentum: 0.000000
2023-10-06 10:10:14,273 epoch 8 - iter 60/304 - loss 0.02596985 - time (sec): 24.76 - samples/sec: 261.20 - lr: 0.000050 - momentum: 0.000000
2023-10-06 10:10:26,103 epoch 8 - iter 90/304 - loss 0.02185886 - time (sec): 36.59 - samples/sec: 258.16 - lr: 0.000048 - momentum: 0.000000
2023-10-06 10:10:37,817 epoch 8 - iter 120/304 - loss 0.02907262 - time (sec): 48.30 - samples/sec: 255.92 - lr: 0.000046 - momentum: 0.000000
2023-10-06 10:10:49,679 epoch 8 - iter 150/304 - loss 0.02586838 - time (sec): 60.16 - samples/sec: 255.13 - lr: 0.000045 - momentum: 0.000000
2023-10-06 10:11:02,166 epoch 8 - iter 180/304 - loss 0.03153234 - time (sec): 72.65 - samples/sec: 256.35 - lr: 0.000043 - momentum: 0.000000
2023-10-06 10:11:12,938 epoch 8 - iter 210/304 - loss 0.03238097 - time (sec): 83.42 - samples/sec: 253.75 - lr: 0.000041 - momentum: 0.000000
2023-10-06 10:11:25,085 epoch 8 - iter 240/304 - loss 0.03115189 - time (sec): 95.57 - samples/sec: 254.39 - lr: 0.000039 - momentum: 0.000000
2023-10-06 10:11:37,295 epoch 8 - iter 270/304 - loss 0.03069472 - time (sec): 107.78 - samples/sec: 254.97 - lr: 0.000038 - momentum: 0.000000
2023-10-06 10:11:49,510 epoch 8 - iter 300/304 - loss 0.03294117 - time (sec): 119.99 - samples/sec: 255.33 - lr: 0.000036 - momentum: 0.000000
2023-10-06 10:11:50,987 ----------------------------------------------------------------------------------------------------
2023-10-06 10:11:50,987 EPOCH 8 done: loss 0.0326 - lr: 0.000036
2023-10-06 10:11:58,982 DEV : loss 0.16147054731845856 - f1-score (micro avg) 0.842
2023-10-06 10:11:58,990 saving best model
2023-10-06 10:12:03,319 ----------------------------------------------------------------------------------------------------
2023-10-06 10:12:14,831 epoch 9 - iter 30/304 - loss 0.03375994 - time (sec): 11.51 - samples/sec: 265.76 - lr: 0.000034 - momentum: 0.000000
2023-10-06 10:12:26,424 epoch 9 - iter 60/304 - loss 0.03249143 - time (sec): 23.10 - samples/sec: 268.41 - lr: 0.000032 - momentum: 0.000000
2023-10-06 10:12:37,836 epoch 9 - iter 90/304 - loss 0.03227701 - time (sec): 34.51 - samples/sec: 268.12 - lr: 0.000030 - momentum: 0.000000
2023-10-06 10:12:49,318 epoch 9 - iter 120/304 - loss 0.03168926 - time (sec): 46.00 - samples/sec: 269.93 - lr: 0.000029 - momentum: 0.000000
2023-10-06 10:12:59,865 epoch 9 - iter 150/304 - loss 0.02780577 - time (sec): 56.54 - samples/sec: 267.92 - lr: 0.000027 - momentum: 0.000000
2023-10-06 10:13:11,363 epoch 9 - iter 180/304 - loss 0.02743918 - time (sec): 68.04 - samples/sec: 269.23 - lr: 0.000025 - momentum: 0.000000
2023-10-06 10:13:22,294 epoch 9 - iter 210/304 - loss 0.02537944 - time (sec): 78.97 - samples/sec: 269.91 - lr: 0.000023 - momentum: 0.000000
2023-10-06 10:13:33,769 epoch 9 - iter 240/304 - loss 0.02592961 - time (sec): 90.45 - samples/sec: 270.22 - lr: 0.000022 - momentum: 0.000000
2023-10-06 10:13:44,979 epoch 9 - iter 270/304 - loss 0.03003227 - time (sec): 101.66 - samples/sec: 270.14 - lr: 0.000020 - momentum: 0.000000
2023-10-06 10:13:56,535 epoch 9 - iter 300/304 - loss 0.02736023 - time (sec): 113.21 - samples/sec: 270.09 - lr: 0.000018 - momentum: 0.000000
2023-10-06 10:13:57,990 ----------------------------------------------------------------------------------------------------
2023-10-06 10:13:57,990 EPOCH 9 done: loss 0.0270 - lr: 0.000018
2023-10-06 10:14:05,126 DEV : loss 0.1611461490392685 - f1-score (micro avg) 0.838
2023-10-06 10:14:05,135 ----------------------------------------------------------------------------------------------------
2023-10-06 10:14:16,176 epoch 10 - iter 30/304 - loss 0.05830548 - time (sec): 11.04 - samples/sec: 263.86 - lr: 0.000016 - momentum: 0.000000
2023-10-06 10:14:26,762 epoch 10 - iter 60/304 - loss 0.03839883 - time (sec): 21.63 - samples/sec: 259.36 - lr: 0.000014 - momentum: 0.000000
2023-10-06 10:14:38,590 epoch 10 - iter 90/304 - loss 0.02785645 - time (sec): 33.45 - samples/sec: 266.82 - lr: 0.000013 - momentum: 0.000000
2023-10-06 10:14:50,176 epoch 10 - iter 120/304 - loss 0.02580621 - time (sec): 45.04 - samples/sec: 270.00 - lr: 0.000011 - momentum: 0.000000
2023-10-06 10:15:01,694 epoch 10 - iter 150/304 - loss 0.02166950 - time (sec): 56.56 - samples/sec: 269.72 - lr: 0.000009 - momentum: 0.000000
2023-10-06 10:15:13,131 epoch 10 - iter 180/304 - loss 0.02232735 - time (sec): 67.99 - samples/sec: 270.60 - lr: 0.000007 - momentum: 0.000000
2023-10-06 10:15:24,968 epoch 10 - iter 210/304 - loss 0.02484298 - time (sec): 79.83 - samples/sec: 271.22 - lr: 0.000006 - momentum: 0.000000
2023-10-06 10:15:36,078 epoch 10 - iter 240/304 - loss 0.02347145 - time (sec): 90.94 - samples/sec: 270.75 - lr: 0.000004 - momentum: 0.000000
2023-10-06 10:15:47,348 epoch 10 - iter 270/304 - loss 0.02655606 - time (sec): 102.21 - samples/sec: 270.85 - lr: 0.000002 - momentum: 0.000000
2023-10-06 10:15:58,494 epoch 10 - iter 300/304 - loss 0.02544801 - time (sec): 113.36 - samples/sec: 270.31 - lr: 0.000000 - momentum: 0.000000
2023-10-06 10:15:59,811 ----------------------------------------------------------------------------------------------------
2023-10-06 10:15:59,811 EPOCH 10 done: loss 0.0252 - lr: 0.000000
2023-10-06 10:16:07,059 DEV : loss 0.16277427971363068 - f1-score (micro avg) 0.8367
2023-10-06 10:16:07,919 ----------------------------------------------------------------------------------------------------
2023-10-06 10:16:07,921 Loading model from best epoch ...
2023-10-06 10:16:10,830 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-06 10:16:17,666
Results:
- F-score (micro) 0.7995
- F-score (macro) 0.5565
- Accuracy 0.6767
By class:
precision recall f1-score support
scope 0.7595 0.7947 0.7767 151
work 0.7455 0.8632 0.8000 95
pers 0.8241 0.9271 0.8725 96
loc 0.2222 0.6667 0.3333 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7610 0.8420 0.7995 348
macro avg 0.5102 0.6503 0.5565 348
weighted avg 0.7623 0.8420 0.7990 348
2023-10-06 10:16:17,666 ----------------------------------------------------------------------------------------------------
|